Skip to content

Retrieval Augmented Generative Engine with DeepSeek RAGE

License

Notifications You must be signed in to change notification settings

GATERAGE/RAGEmini

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

81 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

RAGEmini

play on RAGE + gemini as MVP MVP to connect with ollama recognizing ollama from localhost if running
MMP to connect with gemini API including experimental from toggle

RAGE folder contains the as yet to be implented Retrieval Augmented Generative Engine
functions as a basic UI for chat response from localhost ollama model or Gemini
with memory.py and logger.py

RAGEmini/
├── src/
│   ├── memory.py
│   ├── logger.py
│   ├── openmind.py
│   └── locallama.py
├── gfx/
│   └── styles.css
├── memory/
│   ├── sessions/
│   ├── knowledge/
│   └── long_term_memory.json
└── rage.py

RAGE + GEMINI == RAGEmini

Retrieval Augmented Generative Engine

RAGE Retrieval Augmented Generative Engine is a dynamic engine designed to learn from context, injest and memory over time.

Context-Aware Responses:

By leveraging the continuously updated data and learning from past interactions, RAGE can understand and respond to nuances in user queries. This ability makes it particularly effective in scenarios where context heavily influences the nature of the response.

Adaptive Response Generation:

As RAGE evolves, it becomes more adept at predicting user needs and adjusting its responses accordingly, ensuring high relevance and personalization
perform manual install or INSTALL not both

manual install

git clone https://github.com/GATERAGE/DeepSeekRAGE
python3.11 -m venv rage
source rage/bin/activate
pip install --no-cache-dir -r requirements.txt
streamlit run rage.py

INSTALL

source install.sh
streamlit run rage.py

About

Retrieval Augmented Generative Engine with DeepSeek RAGE

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 88.8%
  • CSS 8.0%
  • Shell 3.2%